Kelp DAO's rsETH cross-chain bridge was attacked, resulting in the loss of 116,500 rsETH. Here’s a quick rundown of the affected tokens: KERNEL, the token of Kelp DAO, is temporarily at the top of the list. - The bad debt primarily impacts AAVE, EUL, and COMP, with AAVE being hit the hardest. - LayerZero’s security model might have systemic issues, making ZRO and KERNEL problematic protocols. - EIGEN, which is also part of the LSDfi sector. - LDO and FLUID are still under observation. The attacker stole $290 million worth of rsETH, deposited it into lending protocols, and borrowed $236 million in WETH. Aave V3 might be facing nearly $177 million in bad debt. Here’s the detailed breakdown: One of rsETH’s advantages is its ability to circulate across chains, deployed on over a dozen networks. To achieve this, Kelp uses a unified cross-chain accounting system. On April 19, 2026, Beijing time, the hacker launched the first attack on Kelp’s accounting system verification process, stealing approximately 116,500 rsETH, which accounted for 18% of the total circulating supply. Kelp immediately suspended the contract, and rsETH across all networks instantly lost its underlying asset backing. The hacker then took these now worthless rsETH and deposited them into mainstream lending protocols like Aave and Compound, borrowing over $236 million in ETH, successfully cashing out the counterfeit tokens. After cashing out, the hacker left behind nearly $177 million in bad debt on Aave’s books, effectively transferring Kelp’s security risk to the entire lending market. But that’s not the end of it. Aave has a mechanism called Umbrella to cover bad debt. Users can stake their assets (like WETH) into Umbrella to earn extra interest. The catch? If the protocol incurs bad debt, the staked funds are automatically deducted to fill the gap. This means the ones footing the bill for the hacker’s actions are actually the innocent stakers who provided funds in Aave to earn modest interest. In theory, AAVE has smoothly shifted the tail-end risk onto retail investors chasing returns. The most ironic part? The annualized yield on Kelp is only 2% to 3%. Users are taking on disproportionate cost risks for this tiny return, essentially subsidizing a chain of project teams (from EigenLayer to Kelp to AAVE).